metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 44100
- name: transcription
dtype: string
- name: file_name
dtype: string
splits:
- name: train
num_bytes: 3435218908.2738
num_examples: 9969
download_size: 3139652750
dataset_size: 3435218908.2738
configs:
- config_name: default
data_files:
- split: train
path: data/train-*